INDEX

Explanations

phrases related to learning, improvement, and benefit

oai_token-act-pair · gpt-3.5-turbo

expressions that indicate potential benefits or positive outcomes from cooperation and learning

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 redacted

-0.68

 McKenna

-0.64

 recounts

-0.64

oval

-0.63

anamo

-0.62

 Bowen

-0.60

letter

-0.59

 Anniversary

-0.59

unct

-0.58

 recounted

-0.57

POSITIVE LOGITS

 prosper

1.16

 unstoppable

1.02

 flourish

0.98

 improve

0.97

 outper

0.97

 thrive

0.97

 prevail

0.95

 unbeat

0.92

 prosperity

0.92

 success

0.90

Activations Density 0.535%

phrases related to learning, improvement, and benefit

expressions that indicate potential benefits or positive outcomes from cooperation and learning

No Comments

No Known Activations